Cost Analysis of an AT Service Delivery Program

RESNA 28th Annual Conference - Atlanta, Georgia

Frances Harris, PhD1 and Stephen Sprigle, PhD, PT2

1Helen Hayes Hospital & 2Georgia Institute of Technology


Cost analyses are an important methodological tool for outcomes researchers. They may help in the assessment of both the effectiveness and efficiency of service delivery programs. This paper reports results of a 30-month cost analysis study of an assistive technology (AT) service delivery program at Helen Hayes Hospital, an acute rehabilitation hospital, located in New York. An algorithm was developed to assess the costs of providing an AT intervention. The resulting data highlight significant disparities in the way costs are measured within an AT program. Suggestions emerge regarding how these data may be used to aid clinicians and programs to assess their effectiveness and efficiency within an outcomes context.

KEYWORDS Assistive technology, cost analysis, outcomes, economic evaluation


The importance of outcomes research among assistive technologists is well established (1-3). Economic evaluations are an important, but underutilized methodology in AT outcomes research (4). They compare the costs and outcomes of various programs or therapies. The “cost side” of an economic evaluation is referred to as a “cost analysis” or “cost description.” As Fuhrer (1) notes, the myriad challenges inherent in cost assessment deter many practitioners from undertaking them. In addition, the measurement tools of economic evaluations need to be adapted to the specific requirements of assistive technology programs, which vary across medical, vocational, and educational service models (5).

External to AT practice, policy makers and insurers rely on economic evaluations, which have traditionally been carried out within a medical model, to gauge the value of health care technologies and services. As funding sources become ever tighter, it becomes imperative for assistive technologists to both differentiate and clarify their services from other medical technologies and to establish their effectiveness.


An objective of this study was to develop and test a methodology to assess the costs of providing an assistive technology (AT) intervention within an acute rehabilitation hospital setting.


A total of 51 subjects were followed over a 30-month period between June 2002 and December 2004. The development of a cost methodology consisted of two steps: 1) adapting economic concepts to the requirements of an AT service program evaluation and 2) developing an algorithm to determine the costs of providing an AT intervention.

The first task was to adapt definitions of economic cost categories to the needs of an AT service program. The most important category of costs for this study concerned the definition of “direct” and “indirect” costs. “Direct costs” were defined as resources specifically related to providing an intervention, e.g., clinical supplies and equipment used during an intervention. “Indirect” costs were defined as the “hidden” costs of an intervention - those that are typically non-billable within the medical payment model, e.g., time spent in writing letters of justification, providing documentation to insurers, visiting schools, and researching solutions.

The second task was to develop an algorithm to calculate both the direct and indirect costs of providing an intervention. First, we used hospital budget accounting reports (that assign costs to the department based on square footage) to determine overhead costs. Then clinician salaries and fringe were assessed and added to overhead costs for total department operational costs (DOC). DOC are “fixed” costs, that is, they are costs that do not vary in the short-term.

Table 1: Assessment of Departmental Costs (2001-2002)
DOC $624,109
DDS $116,264
IDS $ 64,502
Total Departmental Costs $804,875

The remainder of costs described here may be described as “variable costs” and included both direct and indirect types. These costs varied directly and proportionately with changes in the volume of services provided – such as office supplies. Direct departmental support costs, for example, reflected that portion of a shop technician’s salary (based on daily time and billing sheets) that directly supported a clinician during an intervention. Indirect departmental support costs consisted primarily of time spent by the director in management and fiscal oversight activities. Office supplies and equipment also were divided into direct and indirect costs. For example, direct equipment costs included pressure mapping equipment used during patient evaluations, while indirect equipment costs included fax machines. Direct departmental support and supplies/ equipment costs were added together (DDS). A similar calculation was performed for indirect departmental support and indirect supplies/equipment. (IDC)

CRT clinicians routinely completed daily time and billing sheets noting both the service provided and the number of service units spent (1 SU = 15 minutes). It is important to note that both direct (billable) (DSU) and indirect (non billable) (ISU) service units were recorded daily. The number of direct and indirect service units was totaled for the period of one year (TSU).

Table 2: Direct and Indirect Service Units for One Year (2001-2002): (1 SU = 15 minutes)
Direct Service Units Indirect Service Units Total Service Units
10,028 17,132 27,160


Using the above data both a direct unit and indirect cost unit were calculated:

Equation 1:

Calculation of Direct (DU) and Indirect Unit (IU) Costs

(DOC/total units) + (DDS/direct units) = DU Cost
$624,109/27,160) + ($116,264/10,028) = $34.57 DU Cost

(DOC/total units) + (IDS/indirect units) = IU Cost
($624,109/27,160) + ($64,502/17,132) = $26.74 IU Cost


Based on cost data supplied by the hospital, a $34.57 direct cost rate and $26.74 indirect cost rate per service unit was calculated for interventions begun in 2002. Between June 2002 and December 2004 a total of 51 subjects were consented: 40 received seating and mobility, 6 augmentative communication and 5 computer access interventions. Cost data based on the number of direct and indirect service units were collected and tabulated by clinician, diagnosis, and equipment.

The mean number of direct units per intervention was 19.96 (st dev=15.55), while the mean number of indirect units was 26.67 (st dev= 20.44). These service units correspond to a mean direct cost per intervention of $690 and a mean indirect cost of $713. The total amount billed by the hospital for AT services was $33,058, representing 94% of the incurred direct costs. When summing direct and indirect costs, billing costs captured 46% of AT intervention costs. Across all 51 subjects, billed amount exceeded direct costs in 29 subjects, but the billed amount did not exceed total intervention costs in any intervention.

Table 3: AT Intervention Costs Between 6/2002-12/2004 for 51 Subjects
Total DSU Mean DSU Total ISU Mean ISU TotalDir Cost Mean DirectCost TotalIndirectCost MeanIndirectCost Total Dir&IndCosts TotalBilled &Dir Costs Billed %TotDir&IndCostsBilled
1018 19.96 1360 26.67 $35,192.26 $690.04 $36,366.40 $713.07 $71,558.66 $33,058 93.94% 46.20%


The results indicated that, on average, direct interventions lasted approximately 4 hours, while indirect activities lasted over 5 hours. Further analysis showed that 12 subjects (24%) required over 10 hours in indirect time. Billable rates are supposed to reflect indirect time but this data showed that costs exceeded billed amounts for all 51 subjects. A program in which more than half of clinical activities are spent on non-reimbursable activities jeopardizes the fiscal health of that program. This suggests both the need for improved program efficiency and continued study of those indirect activities that comprise the majority of intervention costs.

And while the majority of direct costs were roughly equal to billed amounts, exceptions were significant. The data support two major reasons for direct costs exceeding billed amounts. In the first instance, subjects were inpatients and institutional billing practices excluded specific charges for AT services. Therefore, subjects are effectively not billed and so do not contribute to departmental effectiveness, explicitly. There may be inefficiencies reflected at the overall institutional level although no data have been collected to examine this. In the second instance, services provided to vocational rehabilitation clients in New York State (VESID) are reimbursed at a flat rate. VESID evaluations are time intensive and may incur many more service units due to travel and meetings with clients, parents, and school personnel. The intent of flat rate reimbursement is to achieve a balance between low, moderate and high effort clients. Within these subjects, VESID clients tended to be complex and time-intensive.

It is critical for service providers to understand that indirect costs have a substantial negative impact on the effectiveness and fiscal health of AT programs. Simply tracking a clinician’s direct and indirect activity can provide valuable feedback to the clinician as well as administration. Cost analysis provides a methodological basis upon which to establish a comprehensive view of program resources. This suggests that program efficiency may be a strong subject for evaluation and review. Additionally, cost analyses, by establishing a large set of data, may be used to compare cost ratios across programs and determine realistic expectations for program efficiency. In so doing, this will effectively establish a baseline upon which to measure performance efficiencies.


  1. Fuhrer, M. (2001). Assistive technology outcomes research: Challenges met and yet unmet. American Journal of Physical Medicine & Rehabilitation, 80(7), 528-535.
  2. Scherer, M., Galvin, J. (1996). An outcomes perspective of quality pathways to the most appropriate technology. In J. Galvin & Mr. Scherer (Eds.), Evaluating, selecting and using appropriate assistive technology (pp. 1-26). Gaithersburg, MD:Aspen.
  3. Smith, R. (1996). Measuring the outcomes of assistive technology: Challenge and innovation. Assistive Technology, 8, 71-81.
  4. Harris F, Sprigle S. (2003) Cost analyses in assistive technology research. Assistive Technology. 15(1):16-27.
  5. Jutai, J., Ladak, N., Schuller, R., Naumann, S., & Wright, V. (1996). Outcomes measurement of assistive technologies: an institutional case study. Assistive Technology, 8(2), 110-120.


This study was funded by The Langeloth Foundation and the National Institute on Disability and Rehabilitation Research, grant number H133A010403. The opinions contained in this paper are those of the grantee and do not necessarily reflect those of the NIDRR and U.S. Department of Education.

Author Contact Information:

Frances Harris, PhD
Center for Rehabilitation Technology
Helen Hayes Hospital
Route 9W
West Haverstraw, NY 10993
Office phone (845) 786-4808